If I showed you the Mona Lisa, you’d likely picture Da Vinci’s masterpiece. But what if I asked 256 people to draw it *from memory* with a mouse? And what if you combined all those wonky, rushed sketches into one image? The result isn’t chaos—it’s something eerily beautiful. That’s exactly what this Reddit user (a game developer) did, and the experiment is a masterclass in how human creativity and data collide in unexpected ways.
## How This Project Happened
The developer built a game where players try to copy famous paintings *as fast and accurately as possible* using a mouse. Over time, they collected hundreds of attempts. When they overlaid all 256 copies of the Mona Lisa and averaged their pixels, the ghostly composite image you see above emerged. It’s like a psychic echo of Da Vinci’s work, shaped by the collective memory—and shaky hands—of regular people.
## What the Averaged Mona Lisa Shows Us
Individually, each sketch is rough. A crooked chin here, a lopsided smile there. But together? The mistakes cancel out, and core details emerge: the smoky expression, the arched eyebrows, even the far-off background. The creator compares it to Jason Salavon’s work, where he averaged pornographic photos or Miss America contestants to explore visual patterns in culture. Both projects highlight how boring or flawed individual efforts can blend into something meaningful when viewed en masse.
Another image shared with the post arranges the 256 sketches in a grid, ordered by quality. The top-left ones look almost professional; the bottom-right corners devolve into surreal abstractions. That grid alone is a story about human skill variation—but the average image? It’s a commentary on how we all carry a slightly distorted mental image of icons, yet collectively cling to the original.
## Fun Visualization Ideas to Explore
The developer is inviting feedback on where to take this. Here’s what I’d suggest:
– **Time-lapse stroke analysis:** Even cooler than averaging pixels would be replaying all the brushstrokes as a collaborative dance. Imagine 256 people painting simultaneously on screen—some starting with the eyes, others rushing to the background, but all converging on the same result.
– **Accuracy over time:** If players try the same painting repeatedly, how does their average line placement shift? They mention saving stroke data—tracking that across weeks/months per user could reveal learning patterns.
– **’Confidence’ heatmaps:** Which parts of the Mona Lisa are most reliably recreated? The famous eyes? The hands? Which details get obliterated by guesswork and haste? A heatmap where lines overlap could highlight what people *remember*, versus what they rely on imagination for.
## What Other Data Could We Use?
If this project is expanding, here’s what might enhance future visuals:
– **Timing stats:** Do faster drawings lose structure equally across all regions, or do humans preserve *some* details under pressure? Tracking seconds per attempt versus accuracy could reveal shortcuts in collective memory.
– **Segmented strokes:** Do players start with the face? The background? Comparing stroke order across attempts might uncover subconscious routines in human drawing.
– **User tags:** Annotating which players had art training versus none could lead to side-by-side averages. Would trained artists “correct” the blur even within the chaos, or do tools (like a mouse) homogenize skill?)
– **Sound triggers:** Monitoring if clicks/mouse movements correlate with problem areas (e.g. louder crowd when people fumble with the curls in her hair) adds multi-sensory depth to the collective process.
## Final Brushstroke
This isn’t just cool data art—it’s a reminder that our messy human attempts, when aggregated, can form something deeper. The averaged Mona Lisa isn’t about perfection; it’s about how a shared cultural touchstone persists despite our flaws. If you’re into interactive creativity or playful data analysis, this project nails that intersection perfectly. Can’t wait for the game to drop. (Link in comments if you’re curious!)
—
*Fascinated by art + algorithms? Check out Jason Salavon’s portfolio for more projects that explore averages, chaos, and patterns in human behavior.*